Please use this identifier to cite or link to this item:
http://hdl.handle.net/20.500.12386/32452
DC Field | Value | Language |
---|---|---|
dc.contributor.author | CESARE, VALENTINA | en_US |
dc.contributor.author | BECCIANI, Ugo | en_US |
dc.contributor.author | VECCHIATO, Alberto | en_US |
dc.date.accessioned | 2022-07-05T14:56:33Z | - |
dc.date.available | 2022-07-05T14:56:33Z | - |
dc.date.issued | 2022 | en_US |
dc.identifier.uri | http://hdl.handle.net/20.500.12386/32452 | - |
dc.identifier.uri | https://doi.org/10.20371/INAF/TechRep/164 | - |
dc.description.abstract | We ported on the GPU with CUDA the Gaia Astrometric Verification Unit-Global Sphere Reconstruction (AVU-GSR) Parallel Solver. The code aims to find the astrometric parameters of ~10^8 stars in the Milky Way, the attitude and the instrumental settings of the Gaia satellite, and the global parameter 𝛾 of the PPN formalism, by solving a system of linear equations, 𝐀 × 𝒙 = 𝒃, with the LSQR iterative algorithm. The coefficient matrix 𝐀 is large, having ~10^11 × 10^8 elements, and sparse. The CUDA code accelerates ≳ 14 times compared to the current version of the AVU-GSR code, parallelized on the CPU with MPI+OpenMP and in production since 2014. This acceleration factor is ~9.2 times larger than the one obtained with a preliminary GPU porting with OpenACC, equal to ~1.5. We obtained this result by running the codes on the CINECA SuperComputer Marconi100, that has 4 NVIDIA Volta V100 GPUs per node, where the MPI+CUDA application has been recently put in production. This analysis represents a first step to understand the exascale behaviour of a class of applications that follow the same structure of this code, employed in several contexts. In the next months, we plan to run this code on the pre-exascale platform Leonardo of CINECA, with 4 next-generation A100 GPUs per node, to better investigate this behaviour. Computing on highly parallel devices, such as GPUs, might imply a consistent power saving, which might go towards the achievement of a Green Computing milestone. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | O.A. Catania | en_US |
dc.title | The MPI+CUDA Gaia AVU-GSR Parallel Solver in perspective of next-generation Exascale Infrastructures and new Green Computing milestones | en_US |
dc.type | Technical report | - |
dc.relation.medium | ELETTRONICO | - |
dc.relation.numberofpages | 14 | en_US |
dc.type.referee | REF_2 | - |
dc.description.numberofauthors | 3 | en_US |
dc.relation.scientificsector | INF/01 - INFORMATICA | en_US |
dc.type.miur | 298 Altro | - |
dc.relation.series | INAF Technical Reports - Rapporti Tecnici INAF | - |
dc.relation.ercsector | ERC sectors::Physical Sciences and Engineering | en_US |
dc.description.oa | 1 – prodotto con file in versione Open Access (allegare il file al passo 5-Carica) | en_US |
dc.contributor.referee1 | VITELLO, FABIO ROBERTO | en_US |
dc.contributor.referee2 | TAFFONI, Giuliano | en_US |
dc.identifier.report | 164 | * |
item.openairecristype | http://purl.org/coar/resource_type/c_18cf | - |
item.fulltext | With Fulltext | - |
item.grantfulltext | open | - |
item.languageiso639-1 | en | - |
item.openairetype | Technical report | - |
item.cerifentitytype | Publications | - |
crisitem.author.dept | O.A. Catania | - |
crisitem.author.dept | O.A. Catania | - |
crisitem.author.dept | O.A. Torino | - |
crisitem.author.orcid | 0000-0002-4389-8688 | - |
crisitem.author.orcid | 0000-0003-1399-5556 | - |
Appears in Collections: | 4.01 Rapporti tecnici INAF |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Technical_report_Gaia_MPI_CUDA_Valentina_Cesare_et_al.pdf | 1.76 MB | Adobe PDF | View/Open |
Page view(s)
167
checked on Sep 19, 2024
Download(s)
56
checked on Sep 19, 2024
Google ScholarTM
Check
Items in DSpace are published in Open Access, unless otherwise indicated.